In an earlier post about Augmented Reality we spoke about task-based testing. A nice suggestion of us, but what is it and how does it work? Well in searching for literature I didn’t find much. The US Air Force uses it for unit testing their software, but with Augmented Reality it can be used during system testing. But there is an essential in task-based testing, as the name says: it uses tasks to test the system. I’ll explain this in this post.

At first there needs to be an decomposition of the system into different tasks. These tasks contain the task that needs to executed for the system to work correctly. The tasks need to be analyzed. The result of the Task Analysis is a task framework whereas all the tasks are listed. When you categorize all those tasks in tasks that need to be done by the system to be successful you get a prioritized list of tasks, a Mission Essential Task List.

We start with an example. The application is an online booking system for rental cars. The application consists of multiple ways of using it.

• Looking for cars
• Reservation of cars
• Check your reservation
• Compare cars
• Book directly
• Mail the back office
• …

And the lists goes on. These functions can be cut down into smaller pieces, like only the direct reservation of a car. This function can be seen as a task. Same is the mailing of the back office. A comparison of cars consist of multiple tasks, go to “compare car”, select car number 1, select car number 2 or more, compare the cars and even go to reservations. These are separate tasks in de Task List. The system shall be tested according to these tasks and not according to the requirements.

There are different paths through the application to accomplish the task.

These tasks consist of mini end-2-end chains. They run through the whole system. And therefore all different parts of the system. This is how it can be used with Augmented Reality. Because AR consists of 3 testable parts, the integrated part needs to be tested over all these parts. The use of task-based testing is the method of testing is to be recommended!

When executing the tasks while testing it is important to know, as a tester, that the paths to the different tasks are not set. The testers needs to find the path through the system for that task. This path should be logged by the tester when executing the tests. A lot like Exploratory Testing.

Task-Based Testing is a testing method to test systems with mini end-2-end chains within the software and hardware. It is not set up according to requirements but according to the different tasks within the system. It can test the use of the system, not quite usability, but also a part functionality and interoperability.

1. electromozzo Says:
   July 28th, 2009 at 10:48

   I love these stories! Keep making them!

2. MishaPowerauto Says:
   July 29th, 2009 at 07:40

   I rarely comment on blogs but yours I had to stop and say Great Blog!!
   http://www.testingthefuture.net – cool!!!!

3. SergeyNikolaev Says:
   July 30th, 2009 at 11:28

   Hello. I think the article is really interesting. I am even interested in reading more. How soon will you update your blog?

4. Andreas Prins Says:
   July 31st, 2009 at 06:14

   Hey Sergey Nikolaev,

   We try to post every week at least once and sometimes twice. If you want to see more about a specific topic please let us know, we can think and write about it if we know something

5. LenaShopogolik Says:
   August 6th, 2009 at 07:04

   Just wanted to say HI. I found your blog a few days ago on Technorati and have been reading it over the past few days.

6. Sdanektir Says:
   August 6th, 2009 at 16:23

   I don’t get it, what do you mean by the 3rd paragraph?

7. Ewald Roodenrijs Says:
   August 7th, 2009 at 12:12

   @sdanektir With AR you can show this information in real-time.

8. Shrav Says:
   December 7th, 2010 at 07:22

   Nice blog.. How can AR be used in software testing? How is it beneficial?